Review a sensitive-event storyboard for necessity, accuracy and likely harm before generating visuals. If a scene adds distress or apparent evidence without improving understanding, remove it or use a less literal explanation.
Define the public-interest purpose
Write one sentence stating what viewers should learn. Keep that separate from goals such as stronger emotional impact or a more dramatic opening.
For an Indian regional-news explainer, distinguish verified event details from general educational context. A generic generated street should not quietly stand in for a specific neighbourhood where something happened. Likewise, invented people should not appear to document actual victims, witnesses or responders.
The editorial question is not simply whether a scene is graphic. A restrained but realistic invented scene can still mislead about what occurred.
Complete a necessity and harm sheet
For each proposed shot, record:
- Learning purpose: What factual or explanatory job does this image perform?
- Evidence basis: Which details are established, and which would be invented?
- Affected people: Who might feel identified, blamed, exposed or misrepresented?
- Intensity: What distressing elements are proposed, and are they necessary?
- Alternative: Could a map, timeline, object or text explain the same point?
- Disclosure: How will viewers recognise reconstruction or illustration?
- Reviewer: Who checks factual accuracy and audience impact before release?
Use short written reasoning rather than a single “safe” score. A low overall score can conceal one unacceptable implication.
Hypothetical example: a flood explainer
Suppose a creator plans a 60-second educational video about flood preparedness for a fictional district. Proposed inputs include a fictional timeline, general preparedness messages and an invented rescue scene involving a distressed family.
The rescue shot is not needed to explain the three intended lessons: recognise warnings, prepare essentials and identify an agreed meeting point. Replace it with clearly illustrative diagrams and a non-identifying household checklist.
Evaluate the revised storyboard by asking viewers what event they believe they saw and which actions they remember. If they infer that the fictional rescue actually happened, the disclosure or imagery still needs work. If they recall the actions without that inference, the explanation is doing its intended job.
Stop before speculative generation
Pause scenes whose evidentiary status or necessity remains unresolved. Next, review a text-only storyboard with an editor and consult the current provider policy. Do not predict acceptance or rejection from the supplied policy excerpt.
When comparing production options in the AI tools catalog, check access format and save terms before buying. Marketplace account/code/token offers are not proof of official subscription equivalence; tool access does not replace editorial review.
Sources and scope
Source material checked 29 October 2026; features/policies may change. Runway's Usage Policy says its policy and terms work to mitigate harm and describes automated systems and internal human review. It does not substantiate a predicted moderation outcome for this storyboard. The necessity sheet and example are editorial advice.